1. Field of the Invention
The present invention relates to a plural camera images capturing and processing apparatus and a composite imaging method which capture an image of a wide area with high resolution.
2. Description of the Related Art
In recent years, structures, such as bridges or tunnels which have been constructed during the high-growth period, have started to reach their expected lifetime. Therefore, there is an increasing demand for inspecting the damaged state of the structures and determining the necessity for restoring or repairing the structures. In the technique according to the related art which inspects the damaged state of the structures, when it is difficult for an inspector to inspect the structure on the road, the inspector observes the structure with the naked eye using the scaffold. However, in recent years, an observation technique using a camera has been introduced in order to reduce costs for the scaffold and to ensure the safety of the inspector.
In the inspection of the damaged state of the structure, the area of an inspection range is large and cracks to be inspected are fine. Therefore, there is a concern that resolution will be insufficient in a single captured image. In the observation technique using the camera, in many cases, the images captured by a plurality of cameras (multiple cameras) are composed to generate a high-resolution image.
As a method of composing the images captured by a plurality of cameras into a high-resolution image, there is a method which searches for a common object from adjacent captured images and connects the images on the basis of the common object.
For example, JP-9-161068-A discloses a method which composes the images captured by a plurality of cameras into one high-resolution image. In JP-9-161068-A, paragraph 0011 discloses that “a laser beam is radiated to an object 2 in the imaging field of view of a CCD camera 1 and the image of the object 2 is captured by the same CCD camera 1”. Paragraph 0012 discloses that “a laser beam is radiated to an object 2 including at least a common imaging portion in the imaging field of view of each of two or more CCD cameras 1 and the image of the object 2 is captured by the CCD cameras 1”. Paragraph 0018 discloses that “laser beams emitted from a laser device 3 are a plurality of spot beams”. Paragraph 0019 discloses that “laser beams emitted from a laser device 3 are two or more line beams which intersect each other”. In JP-9-161068-A, paragraph 0049 discloses that “the field of view of the camera 1 or the laser beam irradiation direction of the laser device 3 is preferably set with allowance, considering the positional deviation of the carriage 10”.
The method which searches for the common object from adjacent captured images and composes the images on the basis of the common object has the problem that a high-speed processing device or a lot of processing time is required. In addition, in some cases, the common object is not detected from adjacent captured images. For example, the common object is not detected when there is no characteristic object in the common imaging region (for example, the surface of a white wall) and when the image of an object with a repetitive pattern is captured in the common imaging region and it is difficult to determine a common pattern. In this case, it is difficult to connect adjacent captured images.
According to the technique disclosed in JP-9-161068-A, a plurality of images are connected to each other, using the laser beam (marker) as the common object. Therefore, it is possible to stably compose images, without taking a lot of processing time. However, the laser beam (marker) is included in the composite image and there is a concern that cracks to be inspected will be concealed by the laser beam.